More information about text formats
I read with great interest the study by Atakan et al., (2022) where they summarize existing evidence regarding the effect of high intensity interval training (HIIT) and sprint interval training (SIT) over fat oxidation during sub-maximal intensity exercise. An impaired fat oxidation is a common feature of patients with obesity and type 2 diabetes. Thus, this meta-analysis provides novel information that could be used by physicians and personal trainers to improve the metabolic health of the above mentioned populations. In this rapid response, I discuss several issues regarding data collection process, statistical modelling and interpretation of the reported findings that raised up after a deep analysis of the studies included in this meta-analysis.
The first meta-analysis of this study evaluated the effect of HIIT/SIT over exercise fat oxidation, summarizing the data from nine studies (Fig. 1). The findings of this meta-analysis are reported in g/min, nevertheless, Arad et al. (60) reported fat oxidation (Fox) relative to fat-free mass (mg∙kg FFM-1∙ min-1), Nybo et al. (30) reported Fox in kJ/min while Zapata-Lamana et al. (64) reported the relative contribution of lipids to energy expenditure. Of note, the authors do not report to request the fat oxidation in g/min from these studies (see data extraction section). Then, ¿How did the authors computed Fox in g∙min-1?
Otherwise, Astorino et al. (44) and Schubert et al (33), evaluated the effect of both HIIT a...
Otherwise, Astorino et al. (44) and Schubert et al (33), evaluated the effect of both HIIT and SIT over fat oxidation. According to the study selection criteria employed by the authors, data from both training regimes must be included in the meta-analysis, giving a total of 11 data for the forest plot. Nevertheless, only the data from 9 studies is reported.
With regard to the first analysis reported in this study (Fig. 1), the authors stated that a random effects meta-analysis was performed. According to the description of the reported data “, a positive value indicates a larger increase in fat oxidation as a result of interval training compared with no exercise”, which suggest that authors employed the inverse-variance approach (not specified by the authors). If this is true, the forest plot suggest that 8/9 studies reported a larger increment of Fox in the experimental group when compared to observed modifications in the control group. Nevertheless, from 9 studies investigating the effects of HIIT/SIT over Fox (including 6 studies that evaluated maximal fat oxidation, MFO), only Jabbour and Lancu (47) and Zapata-Lamana et al. (64) reported a significant increment of Fox. Thus, there is a disagreement between the data reported by the studies and the data presented in the meta-analysis. Moreover, because Fox is reported in different units across studies, authors must employ standardized mean difference instead of mean difference. In addition, three of the included studies (44, 56, 64) performed the assessments of fat oxidation on three or more separate occasions. However, the authors do not describe how did they adjusted for repeated measures/multiple comparisons in their analysis. In this sense it would be more appropriate to employ the reported effect size reported or computed by each study (Cohens d for studies that performed only 2 measurements and n2p for studies that performed 3 measurements).
In agreement with the authors, criteria for interpreting whether the enhancement of Fox was substantial or poor remains undefined. However, from the analytical point of view, the observed increment of Fox must be higher than inter-day Fox variation. For example, previous studies report that MFO shows an inter-day variation between 11-26%1-3. Thus the increment of MFO must be larger than such variation, otherwise, authors are only reporting the variation in MFO between exercise test performed at baseline and after training intervention.
Finally, I would like to highlight that in spite HIIT/SIT induced a larger increment on fat oxidation rates in comparison to moderate intensity training (MOD), HIIT/SIT is not better for burning fat in comparison to MOD as equivocally interpreted by many readers who shared the findings from this study in different websites (see altmetric). Indeed, enhancing Fox would be useless if people keep training at high intensity (>85% of VO2max or %PPO) where fat oxidation became negligible. Of note, this meta-analysis does not discuss whether HIIT/SIT improves Fox in different intensity domains, including vigorous intensity. On the contrary, the results from this meta-analysis, suggest that HIIT/SIT and MOD must be combined; the HIIT/SIT will increase Fox during MOD which might contribute to reach a negative fat balance and prevent lipid accumulation in the sarcolemma which impairs insulin signaling.
Note: numbers in the parenthesis correspond to the reference number of the studies included in the meta-analysis.
1. Croci I, Borrani F, Byrne NM, et al. Reproducibility of Fatmax and fat oxidation rates during exercise in recreationally trained males [published correction appears in PLoS One. 2014;9(11):e114115. Byrne, Nuala [corrected to Byrne, Nuala M]; Wood, Rachel [corrected to Wood, Rachel E]; Hickman, Ingrid [corrected to Hickman, Ingrid J]]. PLoS One. 2014;9(6):e97930. Published 2014 Jun 2. doi:10.1371/journal.pone.0097930
2. Dandanell S, Præst CB, Søndergård SD, et al. Determination of the exercise intensity that elicits maximal fat oxidation in individuals with obesity. Appl Physiol Nutr Metab. 2017;42(4):405-412. doi:10.1139/apnm-2016-0518
3. Robles-González L, Gutiérrez-Hellín J, Aguilar-Navarro M, et al. Inter-Day Reliability of Resting Metabolic Rate and Maximal Fat Oxidation during Exercise in Healthy Men Using the Ergostik Gas Analyzer. Nutrients. 2021;13(12):4308. Published 2021 Nov 29. doi:10.3390/nu13124308